Inertial measurement units (IMUs) are critical to the proper operation of inertial navigation and guidance systems. Such systems are used on ships, aircraft, spacecraft, etc.
A typical IMU utilizes a cluster of three accelerometers and three gyros mounted to a structure which is shock isolated. The three accelerometers are used to measure linear acceleration while the gyros are used to measure angular rate.
U.S. Pat. No. 4,920,801, the mathematical equations of which are hereby incorporated by reference, is directed to a monolithic accelerometer capable of sensing linear acceleration in three orthogonal directions. The accelerometer utilizes three co-planar cantilever beams. A mass is formed on each of the cantilever beams. The mass is asymmetrically arranged about the neutral plane of the beam such that the sensing axes passes through the mass at an angle with respect to the plane of the beams. The beams are arranged at 120 degrees with respect to one another such that the sensing axes are substantially orthogonal. Although this accelerometer may be used to measure linear acceleration, it cannot measure angular rate.
Sundstrand Data Corporation has pioneered the development of a Single Coriolis Inertial Rate and Acceleration Sensor (SCIRAS.sup..TM.). In an article by Rand Hulsing II entitled "Single Coriolis Inertial Rate and Acceleration Sensor", Journal of the Institute of Navigation, Vol. 35, No. 3, pp. 347-59 (Fall 1988), the inventor describes a proof-of-concept mechanism which was capable of simultaneously measuring both linear acceleration and angular rate with the same accelerometer structure. The mechanism utilized two back-to-back linear accelerometers disposed on opposite sides of a flexible parallelogram structure. The parallelogram structure was rocked at a predetermined dither frequency. As the parallelogram is rocked about flexures at its corners, a predominantly linear motion is applied to both accelerometers in equal and opposite directions. Using the difference between the two accelerometer outputs, the linear component is measure. Using the sum of the two outputs, the linear components cancel, and only the Coriolis components remain. Thus the small Coriolis acceleration signal associated with angular rate can be extracted from a large linear vibration by matching the scale factors of the two accelerometers.